INDEX

Explanations

references to humans or human-related actions

oai_token-act-pair · gpt-3.5-turbo

references to humans and their characteristics

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

arity

-0.71

 Comprehensive

-0.71

RET

-0.67

 Transparency

-0.67

 Scarborough

-0.65

paragraph

-0.64

onite

-0.62

RAG

-0.62

 Buckingham

-0.61

 Statements

-0.61

POSITIVE LOGITS

 beings

1.06

folk

0.98

oids

0.86

 readable

0.79

zee

0.78

izers

0.77

anguages

0.76

 Humans

0.74

 living

0.72

 humans

0.71

Activations Density 0.023%

references to humans or human-related actions

references to humans and their characteristics

No Comments

No Known Activations